A Fast and Precise Plane Segmentation Framework for Indoor Point Clouds

Abstract

To improve the efficiency and accuracy of plane segmentation for indoor point clouds, this paper proposes a fast and precise plane segmentation framework which mainly consists of two steps: plane rough segmentation and precise segmentation. In the rough segmentation stage, the point clouds are firstly voxelized, then the original plane is extracted roughly according to the plane normal vector and nearest voxels conditions. Based on the results of rough segmentation, a further operation composed of downsampling and density-based spatial clustering of applications with noise (DBSCAN) is adopted to produce efficient and precise segmentation. Finally, to correct the over-segmentation, the distance and normal vector angle thresholds between planes are taken into consideration. The experimental results show that the proposed method improves the efficiency and accuracy of indoor point cloud plane segmentation, and the average intersection-over-union (IoU) achieves 0.8653.